Persons infected by Mycobacterium intracellulare, as demonstrated by positive skin test to the antigen, are concentrated in the southeastern coastal area of the United States. This geographical distribution is unlike that of persons exposed to M. tuberculosis and provides a hypothesis to the possible source of infection with M. intracellulare and other atypical mycobacteria. Our research proposes to test the hypothesis that the source of the infection is mycobacteria residing in either the marine or fresh water of the southeastern United States or animals (birds especially) that are associated with marine habitats. These mycobacteria are concentrated at the surface of bubbles formed during wave action on the ocean. Inasmuch as M. intracellulare at a bubble surface becomes concentrated in droplets that are produced when bubbles burst at a seawater surface, these droplets can become a potential source of infection. Since the predominating direction of air flow during portions of the year is inland from the ocean, the droplets would be brought to those living along the coast. The experimental approach involves sampling air as well as surface and subsurface waters in various areas of the southeastern United States, at and off the coast of Virginia and Georgia. Preliminary studies by the principal investigators have already demonstrated the effectiveness of this approach. The concentration of mycobacteria, their frequency of isolation, and identification of each isolate will be part of this study. The data of concentration and frequency of isolation will be compared to samples from areas where the incidence of infection with M. intracellulare is lower (upstate New York).